Robust Group-Level Inference in Neuroimaging Genetic Studies

Abstract : Gene-neuroimaging studies involve high-dimensional data that have a complex statistical structure and that are likely to be contaminated with outliers. Robust, outlier-resistant methods are an alternative to prior outliers removal, which is a difficult task under high-dimensional unsupervised settings. In this work, we consider robust regression and its application to neuroimaging through an example gene-neuroimaging study on a large cohort of 300 subjects. We use randomized brain parcellation to sample a set of adapted low-dimensional spatial models to analyse the data. We combine this approach with robust regression in an analysis method that we show is outperforming state-of-the-art neuroimaging analysis methods.
Document type :
Conference papers
Liste complète des métadonnées

https://hal.inria.fr/hal-00833953
Contributor : Virgile Fritsch <>
Submitted on : Thursday, June 13, 2013 - 5:17:24 PM
Last modification on : Thursday, March 7, 2019 - 3:34:14 PM
Document(s) archivé(s) le : Tuesday, April 4, 2017 - 9:47:43 PM

Files

rlm.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00833953, version 1

Collections

Citation

Virgile Fritsch, Benoit Da Mota, Gaël Varoquaux, Vincent Frouin, Eva Loth, et al.. Robust Group-Level Inference in Neuroimaging Genetic Studies. Pattern Recognition in Neuroimaging, Jun 2013, Philadelphie, United States. ⟨hal-00833953⟩

Share

Metrics

Record views

567

Files downloads

321